Hypothesis Test 1. A consumer group, concerned about the mean fat content of a certain grade of steak burger submits to an independent laboratory a random sample of 12 steak burgers for analysis. The The percentage percentage of fat fat in in each of the steak burgers is as follows: 21 18 19 1 18 2! 22 19 2! 1! 18 1" The manufacturer claims that the mean fat content of this grade of stea steak k burg burger er is less less than than 2#$. 2#$. Assu Assumi ming ng perc percent entag age e fat fat cont conten entt to be norm normal ally ly distributed with a standard de%iation of &, carry out an appropriate hypothesis test in order to ad%ise the consumer group as to the %alidity of the manufacturer's claim
2. (uring a particular week, 1& babies were born in a maternity unit. )art of the standard procedure is to measure the length of the baby. *i%en below is a list of the lengths, in centimeters, of the babies born in this particular week. !9 "# !" "1 !+ !9 !8 "! "& "" !" "# !8. Assuming Assuming that this sample sample came came from from an underlyi underlying ng normal normal populatio population, n, test, at the "$ signicance le%el, the hypothesis that the population mean length is "# cm.
3. A random sample of 12 steel ingots was taken from a production line. The masses, in kilograms, of these ingots are gi%en below. below. 2!.8 .8 28.1 2!.8 2+.! 22.1 2!.+ 2+.& 2+." 2+." 2+.8 2+.8 2&.9 2&.9 2&.2 2&.2 Assum Assuming ing that that this this samp sample le came came from from an an under underlyi lying ng normal normal population, in%estigate the claim that its mean e-ceeds 2".# kg. / 1$
4. A car manufacturer introduces a new method of assembling a particular component. The old method had a mean assembly time of !2 minutes. The manufacturer would like the assembly time to be as short as possible, and so he e-pects the new method to ha%e a smaller mean. A random sample of assembly times 0minutes taken after the new method had become become established established was 2+ &9 28 !1 !+ !2 &" &2 &8 tating tating any necessary necessary distributional assumptions, in%estigate the manufacturer's e-pectation. / 2 $
5. A random sample of 1" workers from a %acuum 3ask assembly line was selected from a large number of such workers. 4%or topwatch, a work5study engineer, asked each of these workers to assemble a one5litre %acuum 3ask at their normal working speed. The times taken, taken, in in seconds, seconds, to compl complete ete these tasks are gi%en gi%en below below:: 1#9.2 1#9.2 1!.2 1!.2 12+.9 12+.9 92.# 1#8." 91.1 1#9.8 11!.9 11".& 99.# 112.8 1.+ 1!1.+ 122. 119.9 Assuming that this sample came from an underlying normal population, in%estigate the claim that the population mean assembly time is less than 2 minutes. / "$
6. A pharmacist claims that more than #$ of all customers simply collect a prescription. 6ne of her assistants notes that, in a random sample of 12 customers, 1# simply collected a pres prescr crip ipti tion on.. (oes (oes this this pro% pro%id ide e su7c su7cie ient nt e%id e%iden ence ce,, at the the "$ le%e le%el, l, to suppo support rt the the pharmacist's claim
7. 4n a sur%ey carried out in un%ille, 1! children out of a random sample of said that they bought the opper comic regularly. Test, at the 1#$ le%el of signicance, the hypothesis that the true proportion of all children who buy this comic regularly is #.&".
8. A random sample of & coee drinkers were each asked to taste5test a new brand of coee. The responses are listed below with ; representing 'like', 4 representing 'indierent', and ( representing 'dislike'. ;(;;(;;;;;(( ;;;;;(;;;;(( ;;;;(;;;;;;( (o these data support the claim that more than half of all coee drinkers like this new brand of coee< / "$
9. )ackets of ground lter coee ha%e a nominal weight of 2## g. The distribution of weights may be assumed to be normal. A random sample of &2 packets had the following weights. 218 2#+ 21! 189 211 2# 2#& 21+ 18& 18 19+ 219 21& 2#+ 21! 2#& 2#! 19" 19+ 21& 212 188 221 21+ 18! 18 21 198 211 21 2## 2#8 4n%estigate the assumption that the mean weight of all packets is 2## g. Test the hypothesis that 1"$ of packets weigh less than 19# g. / &$
10.
=mployees of a rm carrying out motorway maintenance are issued with brightly colored waterproof >ackets. These come in dierent si?es numbered 1 to ". The last !# >ackets issued were of the following si?es. 2 & & 1 & & 2 ! & 2 " ! 1 2 & & 2 ! " & 2 ! ! 1 " & & 2 & & 1 & ! & & 2 " 1 ! ! Assuming that the !# employees may be regarded as a random sample of all employees, test the hypothesis, at the "$ signicance le%el, that !#$ of all employees re@uire si?e &. Test the claim that si?e & is the median si?e. 11. The Acme ompany has de%eloped a new battery. The engineer in charge claims that the new battery will operate continuously for at least + minutes longer than the old battery. To test the claim, the company selects a simple random sample of 1## new batteries and 1## old batteries. The old batteries run continuously for 19# minutes with a standard de%iation of 2# minutesB the new batteries, 2## minutes with a standard de%iation of !# minutes. Test the engineer's claim that the new batteries run at least + minutes longer than the old. Cse a #.#" le%el of signicance. 0Assume that there are no outliers in either sample.
12.
orty5four si-th graders were randomly selected from a sch ool district. Then, they were di%ided into 22 matched pairs, each pair ha%ing e@ual 4D's. 6ne member of each pair was randomly selected to recei%e special training. Then, all of the students were gi%en an 4D test. Test results are summari?ed below.
)air
Training Eo training
1
9"
9#
2
89
8"
&
+
+&
!
92
9#
"
91
9#
"&
"&
+
+
8
8
88
9#
9
+"
+8
1#
8"
89
11
9#
9"
Trainin
Eo
g
training
12
8"
8&
1&
8+
8&
1!
8"
8&
1"
8"
82
1
8
"
1+
81
+9
18
8!
8&
19
+1
#
2#
!
!+
21
+"
++
)air
22 8# 8& (o these results pro%ide e%idence that the special training helped or hurt student performance< Cse an #.#" le%el of signicance. Assume that the mean dierences are appro-imately normally distributed.
13.
Fithin a school district, students were randomly assigned to one of two Gath teachers 5 Grs. mith and Grs. Hones. After the assignment, Grs. mith had students, and Grs. Hones had 2" students. At the end of the year, each class took the same standardi?ed test. Grs. mith's students had an a%erage test score of +8, with a standard de%iation of 1#B and Grs. Hones' students had an a%erage test score of 8", with a standard de%iation of 1".
Test the hypothesis that Grs. mith and Grs. Hones are e@ually eecti%e teachers. Cse a #.1# le%el of signicance. 0Assume that student performance is appro-imately normal.
14.
The =6 of a large electric utility claims that 8# percent of his 1,###,### customers are %ery satised with the ser%ice they recei%e. To test this claim, the local newspaper sur%eyed 1## customers, using simple random sampling. Among the sampled customers, +& percent say they are %ery satisied. ased on these ndings, can we re>ect the =6's hypothesis that 8#$ of the customers are %ery satised< Cse a #.#" le%el of signicance.
15.
The =6 of a large electric utility claims that 8# percent of his 1,###,### customers are %ery satised with the ser%ice they recei%e. To test this claim, the local newspaper sur%eyed 1## customers, using simple random sampling. Among the sampled customers, +& percent say they are %ery satised. ased on these ndings, can we re>ect the =6's hypothesis that 8#$ of the customers are %ery satised< Cse a #.#" le%el of signicance.
16.
uppose the Acme (rug ompany de%elops a new drug, designed to pre%ent colds. The company states that the drug is e@ually eecti%e for men and women. To test this claim, they choose a simple random sample of 1## women and 2## men from a population
of 1##,### %olunteers. At the end of the study, &8$ of the women caught a coldB and "1$ of the men caught a cold. ased on these ndings, can we re>ect the company's claim that the drug is e@ually eecti%e for men and women< Cse a #.#! le%el of signicance.
17.
uppose the pre%ious e-ample is stated a little bit dierently. uppose the Acme (rug ompany de%elops a new drug, designed to pre%ent colds. The company states that the drug is more eecti%e for women than for men. To test this claim, they choose a simple random sample of 1## women and 2## men from a population of 1##,### %olunteers. At the end of the study, &8$ of the women caught a coldB and "1$ of the men caught a cold. ased on these ndings, can we conclude that the drug is more eecti%e for women than for men< Cse a #.#1 le%el of signicance.
18.
Cn diseIador de productos estJ interesado en reducir el tiempo de secado de una pintura. e prueban dos fKrmulas de pinturaB la fKrmula 1 tiene el contenido @uLmico estJndar y la fKrmula 2 tiene un nue%o ingrediente secante @ue tiende a reducir el tiempo de secado. (e la e-periencia se sabe @ue la des%iaciKn estJndar del tiempo de secado es ocho minutos y esta %ariabilidad inherente no debe %erse afectada por adiciKn del nue%o ingrediente. e pintan &" placas con la fKrmula 1 y otras &" con la fKrmula 2. ;os dos tiempos promedio de secado muestrales son 11 minutos para la fKrmula 1 y 112 minutos para la fKrmula 2. MA @uN conclusiKn puede llegar el diseIador del producto sobre la ecacia del nue%o ingrediente, al ni%el de signicancia #,#1<
19.
inco muestras de una sustancia ferrosa se usan para determinar si hay una diferencia entre un anJlisis @uLmico de laboratorio y un anJlisis de 3uorescencia de rayos O del contenido de hierro. ada muestra se di%ide en dos submuestras y se aplican los dos tipos de anJlisis. A continuaciKn, se presentan los datos codicados @ue muestran los anJlisis de contenido de hierro: uponga @ue las poblaciones son normales, pruebe con un ni%el de signicancia de #,#" si los dos mNtodos de anJlisis dan, en promedio, el mismo resultado O5ray hemical
2.# 2.# 2.& 2.1 2.! 2.2 1.9 2." 2.& 2.!
uponga @ue las poblaciones son normales, pruebe con un ni%el de signicancia de #,#" si los dos mNtodos de anJlisis dan, en promedio, el mismo resultado.
20.
Una muestra de "# familias de una comunidad muestra @ue 1# de ellas estJn %iendo
un programa especial de tele%isiKn sobre la economLa nacional. =n una segunda comunidad 1" familias de una muestra aleatoria de "# estJn %iendo el programa especial de tele%isiKn, a continuaciKn se prueba la hipKtesis de @ue la proporciKn general de tele%identes en las dos comunidades no diere, usando el ni%el de signicancia de 1$
21.
e ponen a prueba la enseIan?a de la =stadLstica empleando =-cel y Finstats. )ara determinar si los estudiantes dieren en tNrminos de estar a fa%or de la nue%a enseIan?a se toma una muestra aleatoria de 2# estudiantes por cada paralelo. (el paralelo A 18 estJn a fa%or, en tanto @ue del paralelo estJn a fa%or 1!. M=s posible concluir con un
ni%el de signicaciKn de #,#" @ue los estudiantes @ue estJn a fa%or de la nue%a enseIan?a de la =stadLstica es la misma en los dos paralelos<.
PRUEBA CHI CUDRAD 22.
upongamos @ue un in%estigador estJ interesado en e%aluar la asociaciKn entre @uienes poseen %ehLculo propio y el ni%el socioeconKmico del conductor. on este ob>eto se toma una muestra de conductores a @uienes se clasica en una tabla de asociaciKn, encontrando los siguientes resultados:
Posee !eh"#$ %o p&opio
'i!e% 'i!e% 'i!e% so#ioe#o()* so#ioe#o()* so#ioe#o()* i#o +,-o i#o *eio i#o ,%to
4 E6
8 1" 1& 1 T,+%, I. Tabla de asociaciKn, %alores obser%ados.
28 1!
M)ermiten estos datos armar @ue el uso del cinturKn de seguridad depende del ni%el socioeconKmico< Csar un ni%el de signicaciKn alfa/#,#".
23.
Tomemos como e>emplo la distribuciKn esperada para los indi%iduos de una poblaciKn @ue son clasicados segPn grupo sanguLneo. egPn estudios reali?ados en poblaciKn, se espera @ue dicha distribuciKn, en porcenta>es, sea la siguiente:
/&$po
&e#$e(#i, espe&,,
A A #
2,#$ ,"$ 9,&$ "8,2$
=n una muestra de 1"# donadores de sangre se encontrK la siguiente distribuciKn:
/&$po
&e#$e(#i, o+se&!,,
A A #
! !8 1" 8&
e a>ustan los datos obser%ados a la distribuciKn teKrica< / 1$
24.
Huan GNnde?, director de Gercadeo de Aladino, tiene la responsabilidad de controlar el ni%el de e-istencias para cuatro tipos de automK%il %endidos por la rma. =n el pasado, ha ordenado nue%os automK%iles ba>o la premisa de @ue los cuatro tipos son igualmente populares y la demanda de cada tipo es la misma. in embargo, recientemente las e-istencias se han %uelto mJs difLciles de controlar. Huan considera @ue deberLa probar su
hipKtesis respecto a una demanda uniforme. ;a demanda es uniforme para los cuatro tipos de autos< Tipo de auto
Kia Fiesta Focus Clio
Ventas observadas 15 11 10 12
Ventas esperadas 12 12 12 12
25.
)aty Al%arado es la directora de in%estigaciKn de )laguicidas. =n su proyecto actual )aty debe determinar si e-iste alguna relaciKn entre la clasicaciKn de efecti%idad @ue los consumidores asignan a un nue%o insecticida y el sitio 0urbano o rural en el cual se utili?a. (e los 1## consumidores a @uienes se le aplicK la encuesta, +" %i%Lan en ?onas urbanas y 2" en ?onas rurales. ;a Tabla 1.2 resume las clasicaciones hechas por los consumidores. Clasificación Arriba del promedio
Promedio Debajo del promedio
Urbano fo = 20
Rural fo =11
fo = 40
fo = 8
fo = 15
fo = 6
;a clasicaciKn y la ubicaciKn son independientes<
26.
A Cni%ersity conducted a sur%ey of its recent graduates to collect demographic and health information for future planning purposes as well as to assess students' satisfaction with their undergraduate e-periences. The sur%ey re%ealed that a substantial proportion of students were not engaging in regular e-ercise, many felt their nutrition was poor and a substantial number were smoking. 4n response to a @uestion on regular e-ercise, #$ of all graduates reported getting no regular e-ercise, 2"$ reported e-ercising sporadically and 1"$ reported e-ercising regularly as undergraduates. The ne-t year the Cni%ersity launched a health promotion campaign on campus in an attempt to increase health beha%iors among undergraduates. The program included modules on e-ercise, nutrition and smoking cessation. To e%aluate the impact of the program, the Cni%ersity again sur%eyed graduates and asked the same @uestions. The sur%ey was completed by !+# graduates and the following data were collected on the e-ercise @uestion:
Eumber of tudents
Eo Qegular =-ercise
poradic =-ercise
Qegular =-ercise
2""
12"
9#
ased on the data, is there e%idence of a shift in the distribution of responses to the e-ercise @uestion following the implementation of the health promotion campaign on campus< Qun the test at a "$ le%el of signicance.
27.
The Eational enter for Realth tatistics 0ER pro%ided data on the distribution of weight 0in categories among Americans in 2##2. The distribution was based on specic %alues of body mass inde- 0G4 computed as weight in kilograms o%er height in meters s@uared. Cnderweight was dened as G4S 18.", Eormal weight as G4 between 18." and 2!.9, o%erweight as G4 between 2" and 29.9 and obese as G4 of or greater. Americans in 2##2 were distributed as follows: 2$ Cnderweight, &9$ Eormal Feight, &$ 6%erweight, and 2&$ 6bese. uppose we want to assess whether the distribution of G4 is dierent in the ramingham 6spring sample. Csing data from the n/&,&2 participants who attended the se%enth e-amination of the 6spring in the ramingham Reart tudy we created the G4 categories as dened and obser%ed the following:
Cnderweight Eormal Feight
U )articipants
6%erweight
6bese
G4S18."
G4 18."52!.9
G4 2".#5 29.9
G4 & #
2#
9&2
1&+!
1###
of
et up the hypotheses and determine le%el of signicance "$
28.
4n a prior e-ample we e%aluated data from a sur%ey of uni%ersity graduates which assessed, among other things, how fre@uently they e-ercised. The sur%ey was completed by !+# graduates. 4n the prior e-ample we used the V 2 goodness5of5t test to assess whether there was a shift in the distribution of responses to the e-ercise @uestion following the implementation of a health promotion campaign on campus. Fe specically considered one sample 0all students and compared the obser%ed distribution to the distribution of responses the prior year 0a historical control. uppose we now wish to assess whether there is a relationship between e-ercise on campus and students' li%ing arrangements. As part of the same sur%ey, graduates were asked where they li%ed their senior year. The response options were dormitory, on5campus apartment, o5campus apartment, and at home 0i.e., commuted to and from the uni%ersity. The data are shown below.
Eo porad Qegular ic =-ercise =-ercis e
Qegular =-ercise
(ormitory
&2
28
6n5ampus
+!
!
!2
Apartment
65ampus Apartment
At Rome
Total
11#
2"
1"
&9
"
2""
12"
9#
et up hypotheses and determine le%el of signicance. R#: ;i%ing arrangement and e-ercise are independentB /#.#"
29.
A randomi?ed trial is designed to e%aluate the eecti%eness of a newly de%eloped pain relie%er designed to reduce pain in patients following >oint replacement surgery. The trial compares the new pain relie%er to the pain relie%er currently in use 0called the standard of care. A total of 1## patients undergoing >oint replacement surgery agreed to participate in the trial. )atients were randomly assigned to recei%e either the new pain relie%er or the standard pain relie%er following surgery and were blind to the treatment assignment. efore recei%ing the assigned treatment, patients were asked to rate their pain on a scale of #51# with higher scores indicati%e of more pain. =ach patient was then gi%en the assigned treatment and after minutes was again asked to rate their pain on the same scale. The primary outcome was a reduction in pain of & or more scale points 0dened by clinicians as a clinically meaningful reduction. The following data were obser%ed in the trial.
T&e,t*e(t /&o$p
(
'$*+e& ith Re$#tio( o 3 Poi(ts
'e P,i( Re%ie!e&
"#
2&
t,(,& P,i( Re%ie!e&
"#
11
'$*+e& ith Re$#tio( o 3 Poi(ts
Test whether there was a signicant dierence in the proportions of patients reporting a meaningful reduction 0i.e., a reduction of & or more scale points using a W statistic, as follows. et up hypotheses and determine le%el of signicance "$
30.
=l director de una escuela clasica a los padres en tres categorLas socio5econKmicas segPn su Jrea de residencia y en tres ni%eles de participaciKn en acti%idades escolares. )robar la hipKtesis de @ue no e-iste relaciKn entre el ni%el socio5econKmico y la participaciKn en acti%idades escolares, con un ni%el de signicaciKn del !$ )articipaciK n Eunca 6casional Qegularme nte
Ei%el de ingreso a>o Gedio Alto 28 !8 1 22 " 1! 1+
+!
&
31.
Cna agencia de publicidad intenta determinar la composiciKn demogrJca del mercado para un nue%o producto. eleccionaron al a?ar +" personas de cada uno de " grupos de edad y les presentaron el producto. ;os resultados de la encuesta son los siguientes: Actitud frente al producto ompra frecuente ompra rara %e? Eunca compra
*rupo de edad 18 5 29
5 &9 !# 5 !9 "# 5 "9 # 5 9
12
18
1+
22
&2
18 !"
2" &2
29 29
2! 29
1&
, (esarrolle una tabla de frecuencias obser%adas y esperadas para este problema + alcule el %alor V2 de la muestra. # =stable?ca las hipKtesis nula y alternati%a. i el ni%el de signicancia es #.#1, Mdebe recha?arse la hipKtesis nula<
32.
(espuNs de aIos de traba>ar en una estaciKn de pesado para camiones, He impson siente @ue el peso por camiKn 0en miles de libras sigue una distribuciKn normal. on el ob>eto de probar esta suposiciKn, He recolecta los siguientes datos un lunes y registra el peso de cada camiKn @ue llega a su bJscula. 85 57 60 81 89 63
89 50 63 76
52 86 95 56 61 75 50 66
65 90 78 95 81 50 62 97
77 60 66 60 61 98 79 67
64 57 92 82 53 63 69 54
61 55 77 93
i He usa la prueba de bondad de a>uste de >i5cuadrada para estos datos, M@uN concluye acerca de la distribuciKn del peso de los camiones< 0Cse un ni%el de signicancia de #.1# y asegPrese de establecer la hipKtesis de interNs. 0 Sugerencia: use cinco inter%alos igualmente probables.
33.
=l coordinador de computaciKn en la escuela de administraciKn cree @ue el tiempo @ue un estudiante de posgrado dedica a leer y escribir correos electrKnicos cada dLa de la semana tiene una distribuciKn normal. )ara e-aminar esta opiniKn, el coordinador recolecta datos un miNrcoles y registra el tiempo en minutos @ue cada estudiante gasta es sus correos electrKnicos. Cse la prueba de bondad de a>uste de >i5cuadrada con estos datos, M@uN concluye acerca de la distribuciKn del tiempo dedicado al correo electrKnico< 0Ctilice #.1# para el ni%el de signicancia y estable?ca con claridad sus hipKtesis. 8.2 12.! 1.2 19.! 12.& 1&.9 1!.& 12.!
+.! 9. 12.8 1#. 18. &.& 1".+ 12.8 2#.! 11.& 1#.9 18.! 18.& 19.2 1!.9 1.+ 11.& 18.1 2#.1
22.! 18.!
.2 12.!
8.+
9.+
1".9
1!.&
1.2
.+
18.!
18.8
2#.!
34.
;a siguiente tabla recoge informaciKn sobre una encuesta reali?ada a una muestra de hogares de " comunidades diferentes respecto a la audiencia del programa radial XRabla el sabelotodoY. )robar la hipKtesis nula de @ue no e-iste diferencia signicati%a en la audiencia entre los hogares de las " ?onas. / !$
35.
=scuchK el programa Eo escuchK el programa
A
Wonas
(
=
12
1+
8
21
2"
&+
!1
2+
1
=l director de una escuela clasicK a los padres de familia en & categorLas socio5econKmicas y en & ni%eles de participaciKn en acti%idades de la escuela. probar la hipKtesis nula de @ue no e-iste relaciKn entre el ni%el socio5econKmico y la participaciKn en las acti%idades escolares. / &$ )articipaciKn en acti%idades Eunca 6casional recuenteme nte
Ei%el socio5 econKmico a>o Gedio Alto 28 !8 1 22 " 1! 1+
+!
&
RE/REI' CRREACI'
1.
;as notas de 12 alumnos de una clase en GatemJticas y Lsica son las siguientes: GatemJticas 2 & ! ! " + + 8 1# 1# Lsica 1 & 2 ! ! ! ! + 9 Rallar el coeciente de correlaciKn de la distribuciKn e interpretarlo.
1#
2. Cna compaILa de seguros considera @ue el nPmero de %ehLculos @ue circulan por una determinada autopista a mJs de 12# kmZh, puede ponerse en funciKn del nPmero de accidentes @ue ocurren en ella. (urante " dLas obtu%o los siguientes resultados:
A##ie(tes ':*e&o e !eh"#$%os a b c d e
3.
"
+
2
1
9
1" 18 1# 8 2# alcula el coeciente de correlaciKn lineal. i ayer se produ>eron accidentes, cuJntos %ehLculos podemos suponer @ue circulaban por la autopista a mJs de 12# km Z h< =s buena la predicciKn< i ayer circularon 1+ %ehLculos a mJs de 12# [mZhora, cuJntos accidentes se esperarLa tener =labore un inter%alo de conan?a para el literal b, utili?ando un ni%el de conan?a del 9&$
=l nPmero de obreros 0en millones ocupados en la agricultura, para los aIos @ue se indican, era: AIo
2##+ 2##8 2##9 2#1# 2#11 2#12 2#1&
2#1 !
6cupados 2,1 2,#! 1,9 1,+! 1,9 1,!9 1,2" 1,1 , )odrLa e-plicarse su e%oluciKn mediante una recta de regresiKn< + DuN limitaciones tendrLan las estimaciones hechas por esa recta< # uJl es el nPmero de traba>adores @ue se esperarLa tener para el aIo 2#1" y 2#1< 4nter%alo de conan?a 09#$ para la estimaciKn del aIo 2#1" e =s able la estimaciKn<
4.
Asocia las rectas de regresiKn y / \- ]1, y / 2- \ 12, y / #,"- ] " a las nubes de puntos siguientes:
Asigna los coecientes de correlaciKn lineal r / #,!, r / \#,8" y r / #,+, a las nubes del problema anterior.
5. ;a tabla siguiente muestra las notas obtenidas por 8 alumnos en un e-amen, las horas de estudio dedicadas a su preparaciKn y las horas @ue %ieron la tele%isiKn los dLas pre%ios al e-amen. Eota
"
+
&
"
8
!
9
Roras de estudio
+
1#
9
!
8
1#
"
1!
Roras de T^
+
2
11
9
&
9
"
, Qepresenta grJcamente los diagramas correspondientes a nota5estudio y nota5T^. + Me obser%a correlaciKn entre las %ariables estudiadas< M(e @uN tipo< M=n @uN caso estimas @ue es mJs fuerte< C Rallar el coeciente de correlaciKn de nota5estudio y nota5T^. MDuN puede deducirse con mJs precisiKn conociendo la nota @ue obtu%o una persona en el e-amen: el tiempo @ue dedicK al estudio o el @ue dedicK a %er la tele%isiKn< Q / #,9!&&82 y #,8!28&. Qespecti%amente. on los mismos datos, hallar las rectas de regresiKn correspondientes y estima para un alumno @ue sacK un 2 en el e-amen: las horas @ue estudiK y ;as horas @ue %io la T^.
6. (urante su primer aIo de %ida han pesado a Garta cada mes. =n la tabla siguiente se dan sus pesos:
Edad
1
2
Peso &,2 &,+
&
!
"
+
8
9
1#
11
12
!,2
",&
",+
,"
,8
+,2
+,9
+,+
8
8,"
, alcula la media y la des%iaciKn tLpica de los pesos.
+ (etermina la ecuaciKn de la recta de regresiKn de y sobre x , e-plicando detalladamente los cJlculos reali?ados y las fKrmulas @ue utili?as Q / a ,22"B 1,+181 b y / #,!8+#] &,#"9#9
# Rallar el coeciente de determinaciKn i Garta llega a tener 1! meses, @uN peso se esperarLa< e i se espera de Garta un peso de +,+ [g. DuN edad deberLa tener< i se espera de Garta un peso de 8,9 [g. DuN edad deberLa tener< ; 6btenga un inter%alo de conan?a 09&$ para la estimaciKn del literal e h 6btenga un inter%alo de conan?a 09!$ para la estimaciKn del literal h
7. =l dueIo de un restaurante de hamburguesas en la ciudad desea determinar la interrelaciKn entre la introducciKn de adere?os importados y las utilidades @ue recibe.
Ctilidades +# (emanda de cJtsup nacional 2 (emanda de cJtsup "# importada
!# 1 "
1## & +"
8# 2
1 !"
1## & &"
on esta informaciKn determinar lo siguiente:
, + #
;a ecuaciKn de regresiKn lineal mPltiple. ;a prueba de signicancia del modelo "$ 4nter%alos de conan?a del 9" $ para los parJmetros del modelo. 4nter%alos de conan?a del 9# $ para la utilidad esperada y la futura cuando la demanda de cJtsup nacional sea de ! y la de cJtsup importada de "#. e =l coeciente de determinaciKn mPltiple. i la demanda de cJtsup nacional es de ! y la importada es de "8B cuJles serLan las utilidades esperadas< ; i la demanda de cJtsup nacional es de & y la importada es de #B cuJles serLan las utilidades esperadas<
8. am Tadeus, dueIo y gerente general de un almacNn de e@uipos electrKnicos, estJ preocupado por el comportamiento de las %entas de un modelo de (^( @ue se %enden en la tienda. e da cuenta de @ue e-isten muchos factores @ue podrLan ayudar a e-plicarlo, pero cree @ue la publicidad y los %endedores son los principales determinantes. am reuniK los siguientes datos: ^entas en unidad es 1
E_ de %endedor *asto en es publicidad ! 1&&
&& 1 +# 82 1+ 2!
& 1# 1& 9
12" 11" 1!# 1 1!" 1!#
, alcule la ecuaciKn de mLnimos cuadrados para predecir las %entas a partir de la + # e
publicidad y los %endedores. i se gasta en publicidad es ` 1&8 y se emplean + %endedores, @uN %entas podrLa pronosticar< i se espera %ender !# e@uipos y se utili?a " %endedores, @uN gasto en publicidad se esperarLa< i se espera %ender !" e@uipo y se gasta ` 18#, cuJntos %endedores se esperarLa tener < alcular el error estJndar de la estimaciKn alcular el inter%alo de conan?a 09#$ para la estimaciKn del literal c
9. A small study is conducted in%ol%ing 1+ infants to in%estigate the association between gestational age at birth, measured in weeks, and birth weight, measured in grams.
, Fe wish to estimate the association between gestational age and infant birth weight. 4n this e-ample, birth weight is the dependent %ariable and gestational age is the independent %ariable.
+ uild a condence inter%al with 9" $, between eight and weight
# 4f the eight is &+.8, which should be the e-pected weight< 4f the weight is 299#, which should be the e-pected eight< e uild an condence inter%al 09#$ for the estimation of c. (etermine the correlation between these %ariables.
10.
The following table shows a researcher works with Al?heimers caregi%ers Table 1. Gade5up data for the predictors of scores for @uality of life.
a) Determie t!e correlatios of t!ese "ariables #it! eac! ot!er$
+ %et&s sa' t!at a perso !as a score o t!e measure of (ocial (upport of 20 ad t!e' !a"e 50)000 i fiacial assets$ *!at #ould 'ou +et if 'ou plu+ t!ese umbers ito t!e re+ressio e,uatio-
. = /$42